Multiple objective optimization of multiple cross-sections to match target beam properties#

Problem description#

The goal and setup are the same as (Single objective optimization of multiple cross-sections to match target beam properties). The only difference is that this example carries out a multi-objective optimization.

Optimization setup#

Method#

Multi-objective genetic algorithm (MOGA) provided by Dakota is used. The method is configured in the following way:

  • Maximum number of functional evaluations: 20,000

  • Size of population: 200

  • Random seed: 1027

The rest are default values given by Dakota.

Running of the example#

  1. Go to {IVABS_ROOT}\examples\e2_uh60_mopt_stf.

  2. Run python run.py uh60_blade.yml.

Result#

Table 12 Performance of the optimization#

Total number of evaluations

20000

Total running time (wall clock)

22183.4 sec (~= 6 hr 10 min)

CPU model

Intel(R) Xeon(R) Gold 6134 CPU @ 3.20GHz

CPU cores

16

Memory

95 GB

../../_images/ivabs_ex_uh60_mopt_stf_result_parallel_coord.png

Figure 8 Parallel coordinates plot of the Pareto front.#